We hypothesize that administration of an inhaled glucocorticoid early in the course of respiratory failure in premature infants will reduce the occurrence of chronic lung disease with few systemic side effects. We hypothesize further that prevention of chronic lung disease, often known as bronchopulmonary dysplasia, in the neonatal period will result in improved pulmonary function and neurodevelopmental status when infants are followed until 18 months of age. These hypotheses will be tested through a multicenter, randomized, double blind, placebo controlled clinical trial. Premature delivery accompanied by elevated inspired oxygen concentration and positive pressure ventilation disrupts the normal process of growth add differentiation of the lung. This disruption is mediated by inflammation, cellular damage and abnormal repair, processes which are mediated through local production and action of cytokines. In this project we will determine whether this process can be inhibited by an inhaled glucocorticoid, beclomethasone, allowing normal growth and differentiation to occur. By starting inhaled steroids as early as 5 days of age we hope to prevent rather than treat chronic lung disease. Direct administration of the drug to its site of action should disrupt local processes of inflammation and damage allowing for lower doses and thus avoiding the complications of high dose systemic steroids. The multicenter approach, involving a large number of patients of diverse background and subtle differences in management, will allow the results to be generalized to the population of infants with respiratory failure as a whole. The major endpoint will be the need for supplemental oxygen and a chest radiograph consistent with chronic bronchopulmonary dysplasia at 28 days of life. The occurrence of these findings at 36 weeks post- conception will also be recorded. Side effects to be compared include growth failure, hypothalamic-pituitary-adrenal axis suppression, hyperglycemia, hypertension and infection. Chronic effects on pulmonary status, growth and neurodevelopment will be determined through follow-up to age 18 months. The trial will be conducted at two Tufts University School of Medicine Neonatal Intensive Care Units, located at New England Medical Center and Baystate Medical Center. The Data Coordinating Center and statistical analysis will be at the Department of Epidemiology and Biostatistics at the Boston University School of Public Health.